Search

A combination of factors including land use change and fire suppression has resulted in the loss of pine savanna habitats across the southeastern U.S., affecting many avian species dependent on these habitats. However, due to the ephemeral nature of the habitat requirements of many pine savanna species (e.g., habitat is only present for a couple of years after a fire), targeted management of such habitats can be challenging. Moreover, the growing numbers of imperiled pine savanna species can make prioritizing management difficult. One potential tool to better inform management of pine savanna species is satellite imagery. Sentinel-2 satellite imagery data provides an instantaneous snapshot of habitat quality at a high resolution and across a large geographic area, which may make it more efficient than traditional, ground-based vegetation surveying. Thus, the objectives of my research were to 1) evaluate the use of remote sensing technology to predict habitat occupancy for pine savanna species, and 2) use satellite imagery-based models to inform multispecies management in a pine savanna habitat. To meet my objectives, I conducted point count surveys and built predictive models for three pine savanna bird species: Bachman's Sparrow (BACS; Peuacea aestivalis), Northern Bobwhite (NOBO; Colinus virginianus), and Red-Cockaded Woodpecker (RCW; Dryobates borealis) across Georgia. I assessed the performance of satellite imagery in predicting habitat occupancy of these pine savanna species and its potential for multispecies management. I found that models created using satellite imagery habitat metric data performed well at predicting the occupancy of all three species as measured by the Area Under the Receiver Operating Characteristic Curve: BACS=0.84, NOBO=0.87, RCW=0.76 (with values between 0.7-1 defined as acceptable or good predictive capacity). For BACS and NOBO, I was able to compare these satellite imagery models to field-based models, and satellite models performed better than those using traditional vegetation survey data (BACS=0.80, NOBO=0.79). Moreover, I found that satellite imagery data provided useful insights into the potential for multispecies management within the pine savanna habitats of Georgia. Finally, I found differences in the habitat selected by BACS, NOBO, and RCW, and that BACS may exhibit spatial variations in habitat use. The results of this study have significant implications for the conservation of pine savanna species, demonstrating that satellite imagery can allow users to build reliable occupancy models and inform multispecies management without intensive vegetation surveying.

Geospatial techniques can used to extract fine-scale spatial data and examine temporal or spatial patterns to inform wildlife conservation planning and management. The overall goal of this thesis was to apply geospatial data and analyses to investigate two systems: conservation of Northern Bobwhite (Colinus virginianus) and harvest management of White-tailed deer(Odocoileus virginianus) in Ohio. Conclusions based on these broad-scale spatial analyses can be used by managers to devise plans which will be actionable and effective for achieving regional population goals. Northern bobwhite populations have been declining in Ohio for decades as a result of habitat loss and degradation caused by successional processes and changes in land use. Landscapes with high juxtaposition and interspersion of early successional, agricultural and forested vegetation are important to fulfill bobwhite resource requirementsthroughout all life stages. I applied land cover composition data to empirically derived distance to cover-type functions with the goal to predict probability of bobwhite occupancy throughout their current range in Ohio. I then compared final model accuracy to a correlational model of naïve landscape indices that similarly predicted occupancy from landscape metrics. Eighty five percent of the study area had a probability of occupancy < 0.25 during both breeding and nonbreeding seasons. This is indicative of inadequate habitat at a regional level, which has been suggested as the most appropriate level of management for this species. I assessed predictive accuracy of both models by predicting occupancy at points where Ohio Division of Wildlife (ODW) whistle count surveys were conducted and comparing predictions to presence or absence of bobwhites. Though both models were accurate to the commonly accepted threshold of 0.7, the distance to cover type model had higher area under the receiver operating curve (AUC) and kappa statistics. The empirical distance to cover type model more accurately distinguished cases of bobwhite presence than the landscape metrics model. This finding could be used to support the value of highly detailed studies done at a fine scale for identifying patterns that can be extrapolated out to scales which are practical and useful for conservation management plans. However, since user accuracy was higher in the distance to cover type model and producer accuracy was higher in the landscape metrics model, context related to the model purpose may be needed to identify which is appropriate in a given situation. Management of white-tailed deer is an essential task for many wildlife management agencies due to their economic, recreational and social importance. Harvest management is a key tool for capturing the benefits and mitigating some detrimental social and ecological impacts of increasingly abundant white-tailed deer populations in Ohio and other midwestern states. I used state-wide survey data of deer hunting events during 2011-2014 to evaluate factors that influenced deer hunter distribution and probability of success within potential Ohio deer management units with the goal to provide important information for harvest managers at a regional scale. While final model results were complex, the strongest relationshipscaptured in all models showed hunters were more likely to hunt but less likely to harvest deer on public compared to private lands. I found differences in final model covariates and the impact they had on hunter use and success between DMUs, which differ based on aspects of human social, geophysical and landcover composition. For example, while all DMUs had a clear trend for hunters to select for locations with a higher percentage of forest and public land, strength of selection for these predictors and which cover types were avoided differed by DMU and,therefore, by landscape context. These results suggest that overall, incentivizing landowners to allow hunting on their property and facilitating access for hunters may be the most effective strategy to increase hunter success. Additionally, information concerning hunter behavior and outcomes in response to spatial variables can be used to devise region-specific management plans to achieve region-specific deer harvest and population goals.

Confronted with threats of habitat loss and climate change, birds that inhabit open-lands such as grasslands and shrublands are among the most rapidly declining groups of birds across North America. One such bird is the Northern bobwhite (Colinus virginianus; hereafter bobwhite) which inhabits heterogeneous landscapes that typically consist of shrubland, grassland, and early-successional forests. Bobwhites are declining at an alarming rate across most of their range, largely due to habitat loss from intense production agriculture. Populations along the northern periphery of bobwhite range are secondarily imperiled by high mortality from severe winter weather. Because bobwhites are a socially and economically important gamebird, there has been great interest from a variety of stakeholders to maintain suitable habitat and promote sustainable populations. As a result of the conservation attention they receive, bobwhite conservation has the potential to benefit non-target songbirds with similar land cover requirements. The umbrella species concept suggests that benefits will accrue to non-target species if viable bobwhite populations are maintained through habitat conservation. The future viability of bobwhite populations is in question in Ohio, so a comprehensive demographic study is needed to estimate probability of population persistence under a variety of future weather and population viability analysis (PVA) scenarios. The first part of my thesis focuses on landscape-scale characteristics that influence bobwhite occupancy in Ohio, and whether occupancy by other open-land species is positively predicted by bobwhite occupancy. The second part of my thesis focuses on projecting population dynamics of bobwhite in Ohio in response to future winter weather and PVA scenarios. I obtained bird survey data from the Ohio Breeding Bird Atlas (2006–2011). I used land cover data (NLCD 2011) to calculate compositional and configurational landscape metrics within a 630 m radius (mean bobwhite home-range size) of breeding bird atlas survey points where bobwhites were detected and not detected. I restricted the extent of my study to 25 counties in southern Ohio that comprise the core of bobwhite range in Ohio. I then constructed single-season occupancy models using the R package `unmarked’ to determine which landscape metrics influenced occupancy after accounting for detection covariates. I predicted occupancy of other open-land species with predicted bobwhite occupancy probability as the sole covariate. I fit additional occupancy models with landscape variables for species whose occupancy was positively predicted (P ≤ 0.05) by bobwhite occupancy. I then compared the fit of bobwhite-only occupancy models with multi-variable landscape models for the same species. A total of 244 individual bobwhites were recorded at 187 unique points among 990 points selected. The highest-ranked bobwhite occupancy model contained the following site variables after controlling for detection: cohesion of forest (β= 0.17, 95% CI: -0.07, 0.41), percent agriculture (β= 0.62, 95% CI: 0.16, 1.07), percent barren (β= 0.13, 95% CI: 0.05, 0.20), and percent grassland (β= 0.13, 95% CI: 0.02, 0.24). Bobwhite occupancy was a significant (P ≤ 0.05) positive predictor for 12 of 35 focal species, and a significant negative predictor for 11 of 35 species. Models with only bobwhite occupancy probability as a predictor was the best supported model for just two species: grasshopper sparrow (Ammodramus savannarum; β= 0.496) and willow flycatcher (Empidonax traillii; β= 0.450). Bivariate plots of cover type space with percent agriculture and amount of forest-agriculture edge as variables revealed that locations occupied by bobwhites contained too much agriculture and not enough edge for shrubland specialists, and insufficient agriculture and too much edge for grassland specialists. These results suggest that bobwhite occupancy is more influenced by percent of open-land cover classes in the landscape. Cover type requirements of bobwhites may be too specialized for effective conservation of an entire suite of species, affording instead conditions that favor a narrower set of species that share similar requirements for composition and configuration of cover types. The broad cover requirements of bobwhites likely result in co-occurrence with several species from different guilds whose needs are partially met by those of bobwhites. I created an integrated population model (IPM) for a declining bobwhite population in southwestern Ohio for the second part of my thesis. The IPM was implemented in a Bayesian framework in program JAGS using the `jagsUI’ package in program R. The IPM combined two sources of count data, five years of winter survival data, and two years of breeding season demographics to quantify the effect of snowfall on winter survival, and to project the population ten years into the future under 4 different winter weather scenarios and 6 different population viability analysis (PVA) scenarios. The IPM estimated that the bobwhite population declined precipitously during 2007–2015 based on fall covey counts (λ= 0.634, 95% CRI: 0.499,0.821) and spring whistle counts (λ= 0.596, 95% CRI: 0.544,0.690). Winter snowfall was confirmed to have a negative effect on winter survival for both adults (β= -0.058, 95% CRI: -0.122, 0.008) and yearlings (β= -0.048, 95% CRI: -0.121, 0.019), and also had the largest influence on population growth rate (λ). Population projections revealed that the bobwhite population had mean yearly extinction probabilities of 0.384–0.410 across all winter weather scenarios. Extinction probability decreased to 0.327 when nest success and summer survival were increased by 20%. However, when winter survival, nest success, and summer survival were all improved by 20%, extinction probability was reduced to 0.287, a ~30% reduction compared with no change in vital rates. Results demonstrated that this population of bobwhites is highly sensitive to effects of severe winter weather, and the population is at a high risk of extinction without management intervention. Improving protective cover to enhance survival during extended periods of snow cover remains necessary to stem the decline of this population, although management efforts should also focus on creating high-quality breeding habitat to improve fecundity and summer survival.

The northern bobwhite (Colinus virginianus) population decline is a severe, rangewide phenomenon beginning >150 years ago and continuing today. In this investigation, I: 1. document the timeline of bobwhite population decline and unintended genetic consequences of attempted remedies, 2) develop a model useful for predicting possible locations of potentially sustainable bobwhite populations in semiarid rangeland in Texas and Oklahoma, and 3) examine the relationship between population monitoring data and meteorological factors. While breeding season call counts of male bobwhite have been used for >70 years to provide estimates of fall populations for hunting, most studies of call counts have focused on mathematics and statistical accuracy of the count, largely overlooking the influence of meteorological factors on call counts. Here, I present the results of >4,400 individual point counts and examine their relationship with meteorological variables recorded at each stop. Humidity was positively correlated with the number of birds recorded (ρ = 0.275, p < 0.001) and temperature was negatively correlated (ρ = -0.252, p < 0.001). The number of birds recorded was significantly higher in wet years than in drought years. There was no significant correlation between wind velocity and number of birds recorded. These results suggest that, while weather does influence call counts and efforts should be made to record meteorological conditions when collecting call count data, the influence of weather may not easily factor into the analysis. These results also provide another line of evidence for decreased breeding behavior during high temperatures. With the increased focus on bobwhite habitat management on a regional scale, there is a need for reliable methods to identify potential bobwhite habitat. To identify bobwhite habitat in semiarid rangeland, I performed classification of LANDSAT scenes of Clay County, Texas from July and December 2015. Stands of mature little bluestem provide excellent bobwhite nesting cover and could be identified using LANDSAT imagery. I scored habitat by type, compared these scores with the results of breeding season call counts from 2014 and 2015 and found significant correlation. When used in combination with other landscape data, this approach can provide a regional context to inform conservation and management decisions.

Northern bobwhites (Colinus virginianus; hereafter, bobwhites) are gallinaceous gamebirds that were once ubiquitous across the eastern United States, but have declined throughout the 20th century. This research was conducted to evaluate the impact of woodlot edge-feathering and land use change on density and survival of bobwhites in Midwestern agricultural landscapes. I examined covey density, survival, and habitat use on 4 private-land study sites in southwestern Ohio to further understanding of winter ecology of bobwhites in relation to habitat characteristics and targeted woodlot edge management. Non-breeding season survival rate was at levels capable of stabilizing the population during the moderate winter of 2012 – 2013 (&Scirc; = 0.393, 95 % CI = 0.215 - 0.596), but was well below the stabilizing rate during the severe winter of 2013 – 2014 (&Scirc; = 0.075, 95 % CI = 0.037 - 0.145). I did not find a relationship between macro- and microhabitat characteristics and weekly survival outcomes during weeks with snow cover, although bobwhites restricted habitat use to areas of high woody stem density with increasing snow depths. This and persistently low survival rates in severe winters suggest that habitat conditions are homogenously poor and are not capable of stabilizing bobwhite populations, even at currently low densities. I examined the predicted probability of use and usable space by examining the proximity of focal habitat types. Loss of early-successional habitat features on the Fee, Thurner, and Wildcat study sites reduced their predicted probability of use and resulted in a net decline in usable space. Targeted successional management of woodlot edges (i.e. edge-feathering) on the Peach Orchard study site increased the predicted probability of use and the proportion of usable space. A reduction in the extent and mean predicted probability of use for herbaceous habitats on Fee indicate that loss of herbaceous habitat resulted in a coarser-grained landscape with less usable herbaceous habitat. Conversely, extent of early-successional herbaceous habitat was reduced on the Wildcat study site, but mean predicted probability of use for the habitat type increased suggesting that sufficient herbaceous cover remained for use during the non-breeding season. Covey densities and usable space increased or decreased concurrently on all study sites; however, changes in covey density and usable space were not proportional. Mean covey density declined between 2009-11 and 2012-14 on Fee (2009-11, 0.59 coveys/ km²; 2012-14, 0.22 coveys/ km²) and Wildcat (2009-11, 1.47 coveys/ km²; 2012-14, 1.37 coveys/ km²), but increased on Peach Orchard (2009-11, 0.45 coveys/ km²; 2012-14, 0.88 coveys/ km²) and Thurner (2009-11, 0.73 coveys/ km²; 2012-14, 0.84 coveys/ km²). Ecological density was calculated as the number of coveys per ha of usable space. Mean ecological density in 2011 and 2014 was 0.031 coveys/ha of usable space (range = 0.017 – 0.044). Management efforts must be prioritized in regions currently occupied by bobwhites and should seek to increase the proportional amount of usable space. Successional management of woodlot edges through edge-feathering and establishment of herbaceous field border programs should be implemented in agricultural landscapes of the Midwest to increase usable space and provide protective cover capable of withstanding severe winters.

Northern bobwhite populations have declined range-wide over the past 40 years. The intensification of agriculture, conversion of pastures to nonnative cool-season grasses, advanced succession, and urbanization have resulted in the deterioration and elimination of bobwhite habitat. Recent conservation efforts have called for a landscape-level approach to the conservation of northern bobwhite populations. However, identifying large areas to manage bobwhite populations is problematic. Reclaimed mine lands offer an opportunity to manage large contiguous tracts of early successional vegetation to conserve northern bobwhite populations. We studied northern bobwhite resource selection throughout the year from August 2009 to March 2014. We investigated the influence of disking, prescribed fire, and herbicide application on bobwhite resource selection. Our study was conducted on Peabody Wildlife Management Area (PWMA), which is a 3,300 ha reclaimed surface mine in Western Kentucky. We used the discrete-choice analysis to compare resource selection on unmanaged and managed units of PWMA. We used locations from 283 bobwhite during the breeding season (1 April – 30 September) and 136 coveys during the non-breeding season (1 October – 31 March). Resource selection on PWMA was influenced most by availability of shrub cover, regardless of season. Bobwhite were found closer to shrub cover than would be expected and selected areas with greater amounts of shrub-open edge density. Similarly, bobwhite selected areas with vegetation characteristics consistent with shrub cover on PWMA, including increased visual obstruction >1 m aboveground and an increased density of woody stems. Management aimed at reducing the density of sericea lespedeza and increasing the openness at ground level positively influenced resource selection. Bobwhite were found closer to disked areas than would be expected and selected areas treated with herbicide to control sericea lespedeza. Bobwhite avoided burned areas during the breeding season but selected burned areas during the non-breeding season. Our results suggest management should focus on increasing the interspersion of shrub cover on reclaimed mine lands. Also, management focused on enhancing the composition and structure of the vegetation (disking and herbicide) should continue. Reclaimed mine land can provide habitat for northern bobwhite, and our results suggest habitat management can improve habitat quality for bobwhite on these lands.

Reclaimed mines present an opportunity to provide large tracts of habitat for northern bobwhite (Colinus virginianus). Reclaimed mine sites are commonly planted to non–native species, including sericea lespedeza (Lespedeza cuneata) and tall fescue (Schedonorus phoenix), which can inhibit growth of more desirable plant species and limit favorable structure for bobwhite. Although bobwhite are found on reclaimed mine sites, there have been no studies documenting how bobwhites use various vegetation types common to reclaimed mine land. Habitat use studies can provide information on how bobwhite select vegetation types on these landscapes and help direct future management decisions. We trapped and radio–marked 841 bobwhite, October 2009 to September 2011, on Peabody Wildlife Management Area, a 3,330 ha reclaimed mine in Kentucky, USA, to investigate how bobwhite used vegetation types and responded to habitat management practices. We used 104 individuals to describe habitat use during the breeding season (1 April–30 September). We found 57 nests and analyzed the movements of 23 brooding adults. We used 51 coveys to describe habitat use during the non–breeding season (1 October–31 March). During the non–breeding season, woody edge was used more than would be expected at random (parameter estimates ≤0.017). During the breeding season, nonbreeding bobwhite used firebreaks dominated by winter wheat and shrub vegetation more than any other vegetation types, and used dense, planted native warm–season grasses (NWSG) and WMA roads least (P

Northern bobwhites are a popular upland game bird species with economic and social importance. Despite nearly a century of research, population declines continue at state and national scales. Precipitous population declines and range-contractions in the Midwest have been caused by land-use changes and suppression of natural disturbance regimes. Reproduction that meets or exceeds annual mortality is essential to sustain viable populations and poor reproductive success is also implicated in declines of grassland bird populations. This study fills an information gap in knowledge of spring dispersal, breeding ecology and behavior, and breeding vital rates of Midwestern bobwhites near the northern limit of their range. It is unclear if recovery of Midwestern bobwhite populations is limited by spring dispersal capabilities and/or breeding vital rates.Radio-telemetry was used to investigate breeding vital rates during 2010 and 2011 on 4 private-lands study sites in southwest Ohio. Bobwhites were outfitted with radio-transmitters with an 8-hr mortality sensor and tracked daily with homing to within ≤20m. Spring dispersal and breeding season vital rates were evaluated and compared to other bobwhite populations to identify potential population-limiting factors associated with spring dispersal and reproduction. Vital rates that met or exceeded range-wide means included spring dispersal distance (1.54 km), nesting effort (re-nest = 26%, double clutch = 65%), male incubation (43% of nests), clutch size (14.4 eggs), egg success (96.9%), and nesting phenology. The mean hatch date for first nest attempts was 1 July, 21 July for male incubated nests, 28 July for renests, 31 July for backdated fall-juveniles, and 24 Aug for double clutches. Breeding season (Apr-Sep) survival (28%) and nest success (27.9%) were lower in southwestern Ohio compared to range-wide averages. Relatively low nest success and breeding season survival indicates that reproduction may limit population growth in this region and that availability of quality breeding habitat is below the threshold needed to offset annual mortality. Providing early successional woody cover within landscapes could improve annual survival and reverse bobwhite population declines in the Midwest. Early successional woody habitat was highly selected at multiple spatial scales during the breeding season. Bobwhites with higher proportions of early succession woody habitat in their nonbreeding home ranges were less likely to disperse or dispersed shorter distance in spring (P = 0.006). Although woody cover was an important component of spring-summer habitat for bobwhites, nest success declined with increasing amounts of woody cover near nest sites (P = 0.05).Continuing support for programs like the Conservation Reserve Program (CRP) that establish permanent, managed grasslands remain important for bobwhite conservation in the Midwest. Vegetation succession coupled with anticipated declines in CRP enrollment would likely diminish usable space for bobwhites in the Midwest during the breeding season and negatively impact nest success. Bobwhites in the agricultural Midwest would benefit from having woody cover in proximity to grassland habitats during the breeding season but woody encroachment within grassland habitats where bobwhites nest should be managed.

Habitat loss due to intensified agriculture, forest maturation, and urbanization is widely accepted as the primary cause of range-wide northern bobwhite (Colinus virginianus) population declines. Although much is known about bobwhite habitat associations, management efforts have failed to halt the decline in most states. Some suggest that bobwhite habitat management has not occurred at the appropriate scale to mitigate the negative impact of changing landscapes. To evaluate the landscape suitability for bobwhites in the core of their range in Ohio, I used radio-telemetry to quantify usable space at various temporal and spatial scales on four private land study sites in southwestern Ohio during 2009-2011. To investigate habitat suitability at a fine-scale, I compared structural and compositional measurements from used and unused winter microhabitats. Mean annual usable space estimates ranged from 10.3 to 24.4 percent for the four sites. Amount of usable space decreased considerably during the non-breeding season due to increased preference for more limited cover types during that period. Covey density ranged from 0.0580-0.2109 coveys per hectare of usable space during the non-breeding season. Analyses of inter-patch distance of cover types suggested that there is a complex interplay of at least three important cover types influencing use on these sites. Predicted probability surfaces showed considerable variability in amount and distribution of areas of high predicted use among sites. Horizontal visual obstruction (less than 1.05 m height) was the most important predictor of microhabitat use across all cover types. Canonical correspondence analysis identified several plant species associated with use. The primary focus of bobwhite management should be increasing usable space based on the composition and inter-patch distance of key cover types. Within landscapes considered suitable for bobwhites in Ohio, habitat management should focus on improving low-level structural complexity, thereby improving the capacity of cover to conceal and protect bobwhites during winter.

Declines in grassland birds have been attributed to loss of habitat, habitat degradation, and changes in land management. In the Mid-South, pasture and hayfield management has focused on maintaining dense stands of non-native forages that do not provide suitable vegetative structure for grassland birds or northern bobwhite. Native warm-season grasses have been promoted for livestock forage and biofuels feedstock. However, little information exists on how these practices affect habitat for grassland songbirds or northern bobwhite in the Mid-South. We conducted a study of two cattle grazing treatments, two hay harvest treatments and a biofuels harvest treatment on vegetative structure for nesting and brood-rearing grassland birds and northern bobwhite in native warm-season grasses. We evaluated vegetative composition and structure during a typical nesting period for grassland songbirds and a typical brood-rearing period for northern bobwhite across Tennessee, 2010 and 2011. We also evaluated invertebrate availability in each grazing treatment. Full-season grazing created suitable structure for nesting and brood-rearing grassland songbirds and northern bobwhite, whereas early-season grazing only provided suitable nesting structure for these species through early summer. Hay and biofuels stands provided adequate nesting cover for grassland songbirds and northern bobwhite, and hay harvests in May and June enhanced structure for brood-rearing northern bobwhite by reducing grass height. However, hay harvests in May or June are likely to impact nesting success for grassland songbirds and northern bobwhite. NWSG planted for biofuels only did not provide suitable structure for northern bobwhite broods. We recommend big bluestem and indiangrass for hay production, as these species mature later and hay harvest is less likely to impact grassland bird reproductive success. In areas where grassland birds and northern bobwhite are a management concern, grazing is a better management tool than haying or biofuels production. We recommend full-season grazing in production stands of native warm-season forages to maximize benefits where grassland birds and northern bobwhite are a management concern.

The northern bobwhite (Colinus virginianus) has experienced range-wide population declines for the past half century. The primary cause has been large-scale habitat loss and fragmentation. Through auspices of the Surface Mining Control and Reclamation Act of 1977 (SMCRA), large tracts of early successional vegetation have been created throughout much of the bobwhite’s range that may be managed to increase usable space. Peabody WMA is a reclaimed coal mine in Western Kentucky where bobwhite have been present in the past. To better understand the dynamics of this population and how habitat on Peabody WMA influences these dynamics, my two objectives were to (1) document survival, cause-specific mortality, and assess multi-scale habitat effects on survival of bobwhite, and (2) estimate nest survival, reproductive efforts, and gauge the effects of habitat composition on these parameters across multiple scales. In relation to bobwhite survival, there was not evidence of multi-scale habitat influence (Part II). Survival increased as the amount of forest vegetation increased within a home range. This was likely related to the availability of woody escape cover associated with forest vegetation on our study site. Pooled seasonal survival rates differed between Ken (S = 0.316, SE = 0.027) and Sinclair (S = 0.141, SE = 0.022) sites. This may have been attributed to differences in habitat suitability or predator abundance. Nest survival rate was low relative to other research (S = 0.317, SE = 0.081; Part III). Nest age was the most influential factor relative to nest survival on our study site and had a positive relationship. Evidence of micro-habitat effects on nest survival existed, though these effects were minimal. Nest survival increased as distance to bare ground increased. This is likely related to the importance of nest concealment on our study site. Our results show that reclaimed mined lands can provide usable space to support bobwhite populations. Management efforts should focus on increasing woody cover within reclaimed vegetation blocks to increase bobwhite survival while increasing the amount of native warm-season grasses for nesting vegetation and litter cover to support reproductive efforts.

Despite a long history of research and accumulated knowledge of factors affecting population growth rates, conservation efforts aimed at reversing population declines of northern bobwhites (Colinus virginianus) have been largely ineffective. Bobwhite population decline and range contraction across the Midwest was driven primarily by changes in land-use practices related to large-scale intensive agriculture, urbanization, and forest succession. Although changes in agricultural landscapes have contributed substantially to the decline, such landscapes remain the only area where active management may improve population growth rates. To understand the ecology of bobwhites in agricultural landscapes in Ohio, I used radio-telemetry to investigate survival and habitat use during the non-breeding season (October–March) on 4 private land study sites in southwestern Ohio during 2009-2011. Known-fates survival estimates were low in both years (Ŝ2009-2010=0.05, 95% CI=0.03, 0.11, Ŝ2010-2011=0.12, 95% CI=0.07, 0.20) and lowest weekly survival coincided with periods of prolonged snow cover. Compositional analysis revealed that coveys used habitat non-randomly at 3 scales; positioning of home ranges within study areas (Λ = 0.320, P < 0.001), positioning of core areas within home range (Λ = 0.599, P = 0.002), and point locations within home ranges (Λ = 0.058, P = 0.002). Early successional woody vegetation (e.g. fencerows and ditches) was the most preferred habitat type at all scales. Differences in selection among study sites revealed that home ranges were preferentially established in areas with grassland cover on 2 agricultural study sites, but not at the site with the greatest amount of grassland cover. Bobwhites mostly used habitat edges such that interior portions of grass and crop fields were used less as they increased in area. Thus small fields with high edge to interior ratios are most advantageous.I tested the influence of habitat use on individual survival by comparing models representing habitat use at 4 spatial scales; site, landscape (78.5 ha buffer), home range, and microhabitat use. Support for different models at 2 scales showed scale-dependent effects of habitat on survival. Higher woody edge density within home ranges was associated with higher survival whereas increasing perennial cover within a 78.5 ha buffer around individuals was related to diminished survival. This relationship suggested that survival was influenced by habitat availability at smaller scales and by factors that affect predator distribution at larger scales. Given the important influence of low non-breeding season survival on population growth rates, management strategies should focus on improving non-breeding season survival in this population by providing woody cover near food sources (e.g. grasslands and row crop fields). Intermediate amounts of grassland and row crop habitat are important in agricultural landscapes, but overrepresentation of any 1 cover type will diminish the suitability of the site for bobwhites during the non-breeding season. Practices focused on improving woody edge habitat along forests are likely to only marginally affect crop production but could significantly improve the amount of usable habitat for bobwhites in agricultural landscapes.

The northern bobwhite (Colinus virginianus) is a gallinaceous upland game bird dependent on early successional grassland habitat for reproduction and survival. Bobwhite populations have been declining range-wide for nearly a half century. The habitat of Babcock-Webb Wildlife Management Area (BWWMA) in southwest Florida is mostly virgin, early successional grassland and pine flatwoods. Although BWWMA is located in the far southern end of the bobwhite range, the area is a popular public land for bobwhite hunting. The BWWMA bobwhite population has declined evidenced by a dramatic decrease in harvest over the last 20 years. The two objectives of my research were to (1) describe nest habitat selection and daily nest survival of the bobwhite population on BWWMA, and (2) evaluate factors related to over-winter (1 October – 30 March) survival of the BWWMA bobwhite population. Specifically, I evaluated nest-site habitat selection and modeled daily nest survival as a function of biologically meaningful spatial, temporal, climatic, and habitat related covariates (Part II). I tested the hypothesis that bobwhites selected nesting habitat at the landscape level. There was no evidence that bobwhites selected specific habitats for nesting, but basin marsh and wet flatwoods cover types were used for nesting slightly more than they were available. The incubation period nest survival rate was 0.477 (SE = 0.027). Daily nest survival rates did not differ among years, the hunting zone in which the nest was located, or between genders of the incubating bird. Nest survival was positively related to the percent of basin marsh habitat within a 1000-m radius of the nest. Daily nest survival declined over the nesting period. I modeled the over-winter survival rates of bobwhites as a function of hunting pressure and other spatial, temporal, climatic and habitat covariates (Part III). The average over-winter survival rate was 0.402 (SE = 0.023). Year, time, and hunting zone were important factors influencing over-winter survival. Hunting pressure was the factor most related to over-winter survival. I evaluated management oriented questions related to over-winter survival of bobwhites on BWWMA. Food strip management and prescribed fire did not appear to be related to over-winter survival. Harvest rates were greater than others reported from studies in the Southeast and results suggested that, to some extent, harvest was additive to natural mortality. If the goal of management is to increase the BWWMA bobwhite population, reduction in harvest rate is one likely effective management strategy for achieving that goal.

Northern bobwhite (Colinus virginianus) are important game birds associated with early successional habitats across the southeastern United States. In the past 30+ years there has been an almost universal decline in bobwhite population numbers despite a long history of management. The Virginia Bobwhite Quail Management Plan was implemented in 1996 to slow and stop the current population declines in Virginia. Virginia Department of Game and Inland Fisheries (VDGIF) personnel identified a lack of knowledge about the broad-scale, landscape level habitats in eastern Virginia. A large scale land cover map along with a detailed understanding of the spatial arrangements of bobwhite habitats will not only aid Virginia's management plan, but also allow focused efforts by our wildlife managers. I explored the possibilities of using remote sensing to map various habitats important to bobwhite. I compared several classification algorithms applied to Landsat TM imagery prior to selecting the classification method that best delineated early successional habitats. After method selection, a classified land cover map for the Coastal Plain and Piedmont of Virginia was generated. Using the classified images available from the first part of the study and 4 years of bobwhite call count data, I studied the landscape level habitat associations of bobwhite. A number of landscape metrics were calculated for the landscapes surrounding bobwhite call count routes and were used in two modeling exercises to differentiate between high and low bobwhite populations. Both pattern recognition (PATREC) and logistic regression models predicted levels of bobwhite abundance satisfactorily for the modeled (84.0% and 96.0% respectively) and independent (64.3% and 57.1%, respectively) data sets. The models were applied to remotely-sensed habitat maps to develop prediction maps expressing the quality of a landscape for supporting a high population of bobwhite based on existing land cover. Finally, I explored the possibility of eliminating the time consuming and very costly step of classifying a remotely-sensed image prior to examining its quality for a particular species. Using raw Landsat TM imagery and bobwhite call count data, I developed predictive logistic regression models expressing the quality of a landscape surrounding a pixel. The first model predicted the probability of the landscape supporting a high bobwhite population. Due to a number of stops with an average of zero, I was also able to generate a model that expressed the probability of the landscape supporting any number of bobwhite. This method also satisfactorily predicted high/low population and presence/absence for the modeled data (65.7% and 83.1%, respectively) and independent data (65.3% and 83.7%, respectively). The method described will allow for rapid assessment of our wildlife resources without having to classify remotely-sensed images into habitat classes prior to analyses.